针对移动机器人的趋光问题,提出了一种基于内发动机机制的控制方法.该方法以生物体感觉运动系统的学习机制为基础,通过评价、行为选择以及取向和决策环节的强化实现对机器人最优趋光控制策略的搜索,使机器人在未知环境下,通过自主的学习和训练,逐渐掌握趋光移动技能.采用马尔科夫定理证明了学习过程的收敛性;仿真实验证明了基于内发动机机制趋光控制方法的有效性;通过与人工势场法的比较,说明了该方法的精确性.
For the mobile robot phototaxis control problem, the control method was proposed based on intrinsic motivation mechanism. According to sensorimotor system learning mechanism the robot achieved the optimal control method through the strengthening links of evaluation, behavioral choices, tropism and decision-making. The robot obtained the independent learning skills in unknown environment and gradually mastered the skills phototaxis through learning and training. The convergence of the algorithm was proved by Markov theorem. Simulation results show the effectiveness of the method. Comparison with the artificial potential method proves the accuracy of this method.